Aurora feels like it is developed by a different team
I could not help to notice the difference in atmosphere between Ariadna and Aurora:
Multiple build a week vs multiple weeks between builds
JetBrains very active on intellij.net vs very few interaction
Almost all new requests were opened or rejected within the day vs hundred of requests are left submitted.
I am not sure if it is a good sign yet. I hope it isn't a sign of rigidity settling in the software. Like some of you responded in one of Thomas thread earlier you are more focused now on the things you are doing.
However as a committed user I can say that I have a lot less incentive to participate than before.
The feedback works both ways. By not engaging your users you miss the opportunity to have a constructive discussion and you run the chance of loosing their interest.
I haven't run any metrics but it definitively feel like there is less activity already, at least meaningful ones.
I understand that you cannot answer all of them in a meaningful way but the effort to open them will be the same now or after. By changing their state now you are giving feedback to your users that yes you have read it and no you do not think you can address it now but you will later. Why having a state Open it isn?t for that?
I suppose you do not intent on leaving hundreds of requests submitted. So it is the difference between a little slower pace now but meaningful feedback and dialog and knowing better where you are or a little faster with losing some of your users and potentially missing some important SCRs (I assume that since you do not open them you are not reading them).
Maybe it is also time to make the intellij.net more scaleable. You could quickly think of
1. Have the SCR categorized based on functional groups: refactoring, cvs, java editing,? You could focus on the ones that you are actively developing in.
2. Have a ranking of users. You could focus on the higher rank first.
As a final thought I will provide a concrete example why I believe it feels like a different team:
The CVS integration is been re-written but none of the VCS Open API requests have been opened/discussed and no discussion AT ALL has been going on on the subject. We have no feedback that you are already working with the knowledge of these requests and it isn?t for the lack of attempts. I know of 3 actively developed VCS plugins that you could leverage off of (p4, subversion, clearcase). To me it looks like the perfect opportunity to get some great feedback on your open API and explain your new direction.
(http://www.intellij.net/forums/thread.jsp?forum=23&thread=26260&tstart=0&trange=100 ).
I am sure you do not intent on raising a wall of silence between your users and yourself but as least to me it feels that way. Maybe the old way cannot work anymore. I could not really guess you handled requests before but it seems that anybody would monitor the forums and the tracker. Maybe you are growing too popular for this. You need a little more structure and dedicate some people time to keep in touch with your user community. As usual rigidity isn?t just in the software. It is also in the organization.
A dedicated IDEA oldtimer.
请先登录再写评论。
On Wed, 07 May 2003 16:14:39 -0400, Gordon Tyler wrote:
i18n management? Hmmm, I'm sure others would be keen on see that plugin.
I know I would.
I think "assigned" state is better because it directly shows the SCR's status. Interpretation of the "Opened with "fixed in"
combination as "assigned" is not so obvious. Furthermore there's a rule that "Fixed in" status cannot be set without the build
number where the issue was resolved.
--
Best regards,
Eugene Zhuravlev
JetBrains, Inc, http://www.intellij.com
"Develop with pleasure!"
"Gordon Tyler" wrote:
Great idea.
Additionally it would be nice if there was some sort of comment if the state
doesn't change from Submitted (Assigned) to anything other after a certain
amount of time. This would not only make the submitter happy because this
shows the request has not been forgotten, but may also help the developer to
get additional information that is needed to process the request - or just get the
confirmation that the problem seems to be solved in a newer build and is
obsolete now.
Sascha
I agree completely. The other problem with the current state is that setting "fix in" is not a state change and ITN does not seem to trigger an email on attribute change (it should) so submitters would never know unless they actively looked.
My comment was just to say that we could do this right now. Obviously you are doing it already because I got a tons of my requests opened and some slated to be fixed in Aurora. This is already the feedback I wanted.
But by all mean implement the ideal solution ;)
that's are great news!!!
Mark Derricutt wrote:
>>The plugin I'm developing is for internal use to help manage i18n of our
>>product, PerformaSure. Does this qualify for "official" plugin developer
>>support?
Unfortunately, it's very product specific and dependent on a custom
logging API.
Briefly, the way it works is it scans the currently open file using PSI
looking for various code constructs that are logging related (e.g. a
private static final LogCategory field and log statements in methods).
From these snippets of code, it constructs a table of the logging
statements in the file which is then displayed in a panel. The developer
can modify the level and message text (the two components we i18n'ize),
and rename the token used to identify the message. The plugin updates
the code and resource bundles accordingly. The plugin can also convert
our pre-i18n logging statements to the new i18n form.
Ciao,
Gordon
--
Gordon Tyler (Software Developer)
Quest Software <http://java.quest.com/>
260 King Street East, Toronto, Ontario M5A 4L5, Canada
Voice: 416-643-4846 | Fax: 416-594-1919
Excellent, now where's that new build ;)
I say let them take their time on this one build before starting the 2 builds a week process. There are 166 bugs fixed so far, and that to me means a much more usable IDE which we can test. If you recall, Eugene (and I like the convenience that 80% of the JetBrain employees are called Eugene) did say that the build is so unstable they're having a hard time using it for internal development... so imagine US! :)
Seems 814 is going to be quite nice to work with, and will give us a nice build to fall back on when the two-a-week kick in.
Just my 2c.
R
An ETA wouldn't hurt, though. Actually, it would fit quite nicely with the
first two points of the "What we will do" list. Seeing so many goodies in
that build, I can't help but check the newsgroups every 15 minutes (I also
wrote a small script that checks the akamai server for a newer version, just
to be sure I don't miss it). They must have some sort of help for this type
of cravings!:-)
Andrei
"Robert S. Sfeir" <robert@codepuccino.com> wrote in message
news:16361384.1052481776422.JavaMail.javamailuser@localhost...
builds a week process. There are 166 bugs fixed so far, and that to me
means a much more usable IDE which we can test. If you recall, Eugene (and
I like the convenience that 80% of the JetBrain employees are called Eugene)
did say that the build is so unstable they're having a hard time using it
for internal development... so imagine US! :)
>
build to fall back on when the two-a-week kick in.
>
"Andrei Oprea" <aoprea@creditwave.com> wrote in message
news:b9ge3k$jfc$1@is.intellij.net...
True, but aren't they on some kind of holliday, binge drinking, vacation
right now? (Then again the bug numbers fixed keep going up over night, so
someone must be sober... or at least I'd hope they are while fixing bugs)
just
type
EAPA? (EAP Anonymous?)
>
>
(and
Eugene)
nice
>
>
+Eugene Belyaev wrote:
Every day there's one person from the team responsible for going through all
submitted requests and assigning them to a particular developer who's most
likely responsible for the problem.+
when do you start with this???
It's the way it always worked.
--
Eugene Belyaev, CTO
JetBrains, Inc
http://www.intellij.com
"Develop with pleasure!"
"Michael Seele" <mseele@guh-software.de> wrote in message
news:16156380.1052730994670.JavaMail.javamailuser@localhost...
all
>